1+ hour, 29+ min ago   (70+ words) 🚀 Announcing Our New AI Model Update! We’re excited to introduce a new update to our AI model, making it smarter, faster, and more efficient at handling operations and requirements. We’re continuously working on improving intelligence, fixing bugs, and enhancing overall…...

13+ hour, 25+ min ago   (566+ words) A coding agent improves its editing tool. Its base model stays exactly the same. Can that count as recursive self-improvement? It can be part of the process. The question is whether the better tool helps the agent produce further improvements,…...

4+ day, 1+ hour ago   (638+ words) Why I compare AI versions scenario by scenario, not average by average Part 9 findings of an experiment: building an LLM-powered support agent with deterministic boundaries. The companion repo contains the full code. My support agent had a known weakness. It…...

4+ day, 18+ hour ago   (792+ words) SWE-bench Verified moved from 60% to nearly 100% in a single year. A benchmark built to separate frontier labs from everyone else stopped doing that job in roughly the time it takes most enterprises to renew a cloud contract. Capability is the…...

4+ day, 18+ hour ago   (734+ words) Gradient Flow | Ben Lorica This is how self improving AI actually starts Self-Improvement Without the Science Fiction A few weeks ago I wrote about AI systems that keep learning after deployment instead of treating every interaction as a fresh start....

2+ week, 5+ day ago   (20+ words) Visualizing AI evals with Inspect Viz Welcome back to our blog series on running,... Tagged with ai, llm, datascience, python....

2+ week, 5+ day ago   (212+ words) View this page in? Domain fit: AI-core · Core AI workload signals detected from paper context and implementation/artifact evidence. Audit each benchmark finding before selecting an implementation path. Evidence refs map to the disclosure below. Evidence graph: 2 refs, 1 links. Utility…...